A New York Times reporter went to South Dakota to report on Kristi Noem's husband. Then the story broke.
New York Times reporter Shawn McCreesh thought there might be a good story to tell about Kristi Noem's husband, Bryon.
A New York Times reporter went to South Dakota to report on Kristi Noem's husband. Then the story broke.
By Peter Kafka
You're currently following this author! Want to unfollow? Unsubscribe via the link in your email.
Chief Correspondent covering media and technology
Gossip about the marriage of former Homeland Security boss Kristi Noem and her husband Bryon Noem, seen here listening to her speak to a Senate committee in 2025, has become a national news story.
ALEX WROBLEWSKI/AFP via Getty Images
2026-04-01T16:03:02.402Z
-
New York Times reporter Shawn McCreesh thought there might be a good story to tell about Kristi Noem's husband, Bryon.
-
That's why McCreesh was already in South Dakota when the Daily Mail ran its bombshell about the Noems.
-
That gave McCreesh a great leg up to write a compelling follow-up.
On Tuesday morning, the Daily Mail published a shocking and salacious story about Bryon Noem, the husband of Kristi Noem, the former head of Homeland Security.
Later that day, The New York Times' Shawn McCreesh published a reaction piece, filled with quotes from the Noems' friends and neighbors, who were astonished to learn that Bryon Noem was allegedly cross-dressing while texting with women who were not his wife. The story is packed with details about the people who live in and around the small town of Castlewood, South Dakota, near the Noem family farm.
That's nowhere near Washington, D.C., where McCreesh is based as a White House reporter. So how did McCreesh find so many people who knew the Noems, so quickly?
Easy: He was already there, talking to them.
McCreesh says he had arrived in South Dakota on Monday, after convincing his editors that there must be an interesting story to tell about Bryon Noem. The state of his marriage to Kristi Noem had become national news, and had been brought up in a congressional hearing, when she was asked if she was having an affair with her advisor Corey Lewandowski. (Noem didn't answer the question, which she criticized as "tabloid garbage.")
"For all the uproar in Washington over the Corey and Kristi saga, there was this big piece of the story just not being examined. And that was the fact of this woman's husband. So I wanted to write about him," McCreesh told me Wednesday, on the phone from a Holiday Inn Express in Watertown, South Dakota.
The story about McCreesh's story is a reminder that going somewhere and talking to the people who live there is a core piece of journalism that can't be automated.
Here's an edited excerpt of my conversation with McCreesh:
Peter Kafka: How did you end up in South Dakota on Tuesday morning, when the Daily Mail story came out?
Shawn McCreesh: I got there on Monday. I thought maybe if I can knock on [Bryon Noem's] door, I could convince him to give me an interview. I wasn't sure what was left of that marriage or what his story was, but I just figured I'd go the old-fashioned way — door-knock and try it out.
If he would talk to me and open up, it could be this amazing story. So I was there for 24 hours talking to townsfolk, friends of the couple, and people who've known them forever.
And then on my second day there, I woke up, and this Daily Mail story dropped, and I was like, "Holy hell." So my story sort of changed.
Shawn McCreesh happened to be in South Dakota for The New York Times when the big story on Kristi Noem's husband broke.
Taylor Hill/FilmMagic
Did you have any inkling about what Bryon Noem was allegedly doing on these chats? Where other people talking about this?
No idea. Hadn't heard of it, didn't know about it. Literally just was interested in writing about Kristi Noem's husband, and this was the week I chose to come.
I hadn't been poking around about it. I hadn't heard any rumors. No one tried to leak me anything. I just thought there's obviously a story here: This woman's running around with her aide all over the country and being asked about their relationship in a congressional hearing under oath. What does the husband think?
Was it a hard sell to get your bosses to fly you out of Washington on a hunch, at a time when there's a lot of news coming out of Washington?
We have a big team, and Times editors are down to be creative and collaborative. And I think that when a reporter gets really excited about a story, they can see that. And luckily, the Times has enough resources, and the Washington bureau is big enough, that I was free to do that. I realize that's a privilege of working at that place, that wouldn't be true in a smaller newsroom.
What was your reaction to the Daily Mail story?
At first, I just thought, "Oh no, I'm blown out of the water here. This is crazy. How can I adapt to this?"
But then I very quickly realized it was kind of an opportunity, and that the news gods had a different plan for me this week. This had suddenly become news, which it wasn't before. It was obviously the topic of all of our conversations yesterday and on everybody's newsfeeds. This subject has become incredibly timely.
And I had already started doing the work, because I'd been making connections with people in the town, and friends of theirs, and understanding this guy in the context of who he is in this town, and also who he is as her husband.
So in a way, I felt I had a head start.
I just retraced my steps from the day before and went back to everywhere I'd been the day before and made lots of phone calls and said, "Well, have you seen this now? And what do you think?" And then banged out a piece on deadline.
Completely unrelated: Last month you asked Donald Trump pointed questions about the bombing of a girls' school in Iran, and the exchange itself became news, and you gained a lot of notoriety. What was that like?
Usually, in a moment involving Trump or politics, when you go viral — and that was about as viral as one can go — if you're being praised, there's an equal but opposite reaction on the other side where the death threats come in and the hate mail and whatever.
In this instance, I was getting mail for weeks. But I never got a single mean email. No tweet, comment, nothing. It just wasn't there. There wasn't another side to that one.
I'd never experienced anything like that. There was nobody who was going to defend him lying about this school bombing.
-
Politics
-
Donald Trump
Read next
Business Insider
https://www.businessinsider.com/new-york-times-bryon-kristi-noem-shawn-mccreesh-interview-2026-4Sign in to highlight and annotate this article

Conversation starters
Daily AI Digest
Get the top 5 AI stories delivered to your inbox every morning.
More about
report
Announcing my retirement to a life of entirely failing to desperately seek renewed meaning
This April 1st, I’m pleased to report that everything is fine. We did it! We saved the world. Congratulations, humanity. There are no more looming apocalypses, no desperate screaming crises, no unendorsedly miserable people on Earth, no creeping degeneration of death and aging existing as a perpetual affront against my values of life and flourishing. Everyone is going to be okay forever, except in the ways they think it’d be interesting and worthwhile to be un-okay. Against all odds, the AI alignment problem has been solved, and more specialized minds than I are managing the tradeoffs involved in steering us towards a vibrant and thriving future. And so in dutiful keeping with the lessons of utopian literature, I shall now descend into the inevitable spiral of ennui and meaninglessness, fo

124x Slower: What PyTorch DataLoader Actually Does at the Kernel Level
<p><strong>TL;DR:</strong> PyTorch's DataLoader can be 50-124x slower than direct tensor indexing for in-memory GPU workloads. We reproduced a real PyTorch issue on an RTX 4090 and traced every CUDA API call and Linux kernel event to find the root cause. The GPU wasn't slow - it was starving. DataLoader workers generated 200,000 CPU context switches and 300,000 page allocations in 40 seconds, leaving the GPU waiting an average of 301ms per data transfer that should take microseconds.</p> <h2> The Problem </h2> <p>A PyTorch user reported that DataLoader was 7-22x slower than direct tensor indexing for a simple MLP inference workload. Even with <code>num_workers=12</code>, <code>pin_memory=True</code>, and <code>prefetch_factor=12</code>, the gap remained massive. GPU utilization sat at 10-2

Day 6/100: Context in Android — The Wrong One Will Leak Your Entire Activity
<p><em>This is Day 6 of my <a href="https://dev.to/hoangshawn/series/37575">100 Days to Senior Android Engineer</a> series. Each post: what I thought I knew → what I actually learned → interview implications.</em></p> <h2> 🔍 The concept </h2> <p><code>Context</code> is one of those Android classes you use dozens of times per day without thinking about it. You pass it to constructors, use it to inflate views, start activities, access resources.</p> <p>But <code>Context</code> isn't one thing — it's a family of related objects with very different lifetimes. Pass the wrong one into a long-lived object, and you've just anchored that object to a screen that the user may have left minutes ago. The screen can't be garbage collected. You've created a memory leak.</p> <p>Not a theoretical one. A r
Knowledge Map
Connected Articles — Knowledge Graph
This article is connected to other articles through shared AI topics and tags.
More in Analyst News
Don’t blame AI for the Iran school bombing | Letters - The Guardian
<a href="https://news.google.com/rss/articles/CBMilwFBVV95cUxNbW9zV3BfcWpaNmxmbDl5cndGTWZ5c1dBMEhpSS01NTRZMFVublhjX09Bdk5GckdTeTN0dEtxSnBRZGlhVmZSTzRmYWFWRkwyQVFyajhIUUJSMkpEVVVvSm0tTVBtT21PdW5WWE1QWEhiVGJsVGJyRkx6T3g4cHN4NFpPYXl5Qk1uM2h5ZHdvUTU2bGh1MDJJ?oc=5" target="_blank">Don’t blame AI for the Iran school bombing | Letters</a> <font color="#6f6f6f">The Guardian</font>

Discussion
Sign in to join the discussion
No comments yet — be the first to share your thoughts!